Research Article: Characteristics of astigmatism in Black South African high school children

Date Published: December , 2017

Publisher: Makerere Medical School

Author(s): Samuel Otabor Wajuihian.


Astigmatism impairs vision at various distances and causes symptoms of asthenopia which negatively impacts reading efficiency.

The aim of conducting this study was to determine the prevalence and distribution of astigmatism and its relationship to gender, age, school grade levels and spherical ametropia.

Using a multi-stage random cluster sampling, 1589 children who included 635 (40%), males, and 954 (60%), females were selected from 13 out of a sample frame of 60 schools. Their ages ranged between 13 and 18 years with a mean of 15.81±1.56 years. The parameters evaluated included visual acuity using the LogMAR chart and refractive errors measured using an autorefractor and then refined subjectively. Axis of astigmatism was presented in the vector method where positive values of J0 indicated with-the-rule, negative values described against-the-rule and J45 represented oblique astigmatism.

The mean cylinder power was −0.09 ± 0.27 and mainly with-the-rule, J0 = 0.01 ± 0.11. The overall prevalence of clinically significant astigmatism (≤ − 0.75 cylinder) in the sample was 3.1% [(95% Confidence interval = 2.1–4.1%)]. Cylinder of at least − 0.25 power was considered to classify astigmatism types. Thus, the estimated distributions of types of astigmatism were: axis- 11.5%, sphero-astigmatism 10.1% and magnitude-astigmatism 11.2% while 67.2% had no cylinder of any magnitude.

The prevalence of astigmatism is relatively low in this population studied. Older children and those in high school grade levels were more likely to have with-the-rule or against-the-rule astigmatism. The prevalence of astigmatism were comparable within but not across regions.

Partial Text

Uncorrected refractive error including astigmatism is a common vision anomaly in school-aged children,1–2 causes visual impairment (VI) and is the second leading cause of treatable blindness worldwide.1–2 Astigmatism occurs when incident light rays from infinity do not converge to a single focal point on the retina due to an irregular curvature of the cornea (corneal astigmatism, obtained from keratometer) or of the lens surfaces.3–6 Refractive astigmatism (that is, total astigmatism) includes both corneal and internal astigmatism and is the type of astigmatism obtained through retinoscopy or subjective refraction.3 Refractive astigmatism may further be classified as sphero-cylinderical (defined in relation to association with spherical component), magnitude of astigmatism (defined according to magnitude or amount) and axis orientation3–4 (defined based on cylinder axis). Astigmatism is clinically important given its associated symptoms which include asthenopia, blurry vision at various viewing distances, spatial distortion in size, shape or inclination of objects and thus degrades image quality.3–4 Furthermore, astigmatism alters emmetropization and is associated with myopia progression and development of amblyopia.5–7 Therefore, knowledge of the distribution of astigmatism in this population will guide clinical practice, research and epidemiology in records and resource allocation.

All data were analyzed by a statistician using the Statistical Package for Social Sciences (SPSS) version 21. Descriptive statistics for visual acuity and refractive error variables were presented with means and standard deviation, and median. The Kolmogorov-Smirnov (K-S) normality test was used to determine normality of the data. The Pearson chi-squared test was applied to determine relationship between age, gender and astigmatism variables as well as association between the vector and traditional methods of representing astigmatism axis. The Pearson correlations test was applied to compare correlations between right and left eye spheres. Logistic regressions were used to examine odd ratios and multivariate associations including relationships between gender, age and astigmatism types. Distributions of variables were presented using tables and proportions and corresponding 95% confidence intervals were presented as estimates of the prevalence.

In this study, astigmatism was analyzed and characterised according to types and in relation to gender, age, and its association with other refractive error in high school children of Zulu ethnic group. The prevalence of clinically significant astigmatism was 3.1%. For the types of astigmatism, with-the-rule, simple myopic and low astigmatism were most frequent. The overall prevalence of astigmatism is lower than findings in some African populations including those by Naidoo et al12 and Kuma et al.14 (Table 6) The differences in findings between this and other studies in African populations12,14 may be related to differences in age and techniques applied to evaluate refractive errors. For example, the authors12,14 studied younger children and reported retinoscopy findings using cycloplegia while findings from present study were derived from subjective refinements of refractive parameters without which may yield lower cylinder power acceptance. However, findings from the present study are comparable to the 3% prevalence estimates reported by Megbelayin et al16 in Nigeria (who also didn’t apply cycloplegia) but higher than results from East African populations19–21 who had participants aged between 12 and 19 years as in present study.

This study characterized astigmatism in a population of Black South African school children of the Zulu, which was not reported previously. The prevalence of clinically significant astigmatism was low and low magnitude, with-the-rule and simple myopic astigmatism were the most frequent types of astigmatism. With-the-rule and against-the-rule astigmatism were more prevalent in the older than younger age groups. The prevalence of astigmatism was comparable within the African regions but differed across regions. Future studies in other racial and ethnic populations in South Africa are recommended so that comparative inferences can be drawn.




Leave a Reply

Your email address will not be published.